I. The Argument
In order to convince the reader about the epistemic or meta-epistemic usefulness of theism for science, we will formulate an argument. Our argument, not drawing on any particular scientific theory but rather based on a more universal meta-scientific reflection, refers to the argument put forward in the late texts of Dummett
(2006, pp. 85–109
; 2010, pp. 39–45
). In our reconstruction and make-up, the latter reads as follows:
I.1. Formulation of the Argument
(D1) The goal of science is to discover “the world as it is in itself” (Dummett 2006, p. 96
) or “to find a description of the physical universe that is independent of our modes of observation” (Dummett 2006, p. 94
(D2) Over time, various sentient beings, including humans, discover the world from different perspectives, and the advancement of science involves moving closer, in cognitive terms, to the true nature of “the world as it is in itself”.
(D3) We are unable to understand to exist
for something in any way other than as “to be possible to perceive” (or to conceive) it (Dummett 2006, p. 97
); thus, “the world as it is in itself cannot be conceived independently of how it is apprehended by any mind” (Dummett 2006, p. 97
(D4) The cognitive subject which corresponds to “the world as it is in itself” needs to be one which conceives it from “[…] no perspective contrasted with other perspectives” (Dummett 2006, p. 96
)—it needs to be a perfect conceiver which can be identified with the omniscient being, i.e., God.
(D5) “The world as it is in itself” is a structure apprehended by God, and this structure determines what can be captured by all sentient beings.
(D6) The goal of science, i.e., to discover “the world as it is in itself”, can be understood only when the existence of God is acknowledged.
Since we are not anti-realists or idealists, we cannot accept the premise (D3) and the (Kantian in content) concept of “the world as it is in itself”.2
Therefore, we propose our own argument, which, assuming the development of knowledge and postulating its purpose or goal, leads to a conclusion of the existence of a subject of perfect knowledge which (at least in epistemological terms) can be identified with God.
The structure of our argument is as follows:
The ultimate goal of practicing science is perfect knowledge.
Science is in the state of development, and, as of now, it barely approximates perfect knowledge.
Perfect knowledge exists:
either as a fictional equivalent of idealization;
or as potential knowledge that can be achieved by people or other finite beings in the future;
or as knowledge currently possessed by God (or a similar, epistemologically infinite being).
(3a) has to be ruled out because scholars actually pursue perfect knowledge, comparing, in a way, their current findings with it and objectivizing (or legitimizing) them in relation to it. In other words, the reality and normativity of perfect knowledge is a necessary condition for scientific endeavours to make sense.
(3b) has to be ruled out on the grounds of the cognitive inexhaustibility of the world and the inability (for finite beings) to obtain the proof of having perfect knowledge.
(3c) has to be assumed because the above (all possible) options (3a) and (3b) are ruled out.
Conclusion: God (or a similar, epistemologically infinite being) exists as the subject of (currently possessed) perfect knowledge.
This argument is elaborated on below.
I.2. The Development and Goal of Science—Premises (1) and (2)
The argument we propose stems from the attempt to identify the goal of practicing science, understood as a comprehensive undertaking, with individual disciplines as its respective elements. This goal, as the general objective of all scientific endeavours, seems to be to obtain well-grounded knowledge about the world (and, in particular, about the so-called physical or external world). We expect scientists to provide knowledge containing true, adequate, and relevant propositions. It means that scientific propositions should:
Give an account of what things are like;
Make it as comprehensive as possible;
Grasp the most fundamental structures of the world to explain and predict phenomena occurring within it and to determine the relationships between them in the context of the world as a whole.
It is obvious that currently we do not have such ultimate or perfect knowledge but are only (to some degree) approximating it. Science is in the state of continual development. The development of science is about:
Eliminating false propositions in favour of true propositions;
Expanding the domains of the world’s fragments, parts, or aspects which are described by true statements;
Discovering (or postulating) ever-more-elemental (or deeper) fundamental structures of the world, as explored by continuously improved scientific theories, i.e., ones with greater explanatory, predictive, and unifying powers (cf. Papineau 1995, pp. 149–52
The final point is of the essence here because the better a theory we have, the more true propositions we can generate for a given research area. After all, the absence of a proper comprehensive theory does not make us have sufficient reasons to claim that propositions currently considered as true actually are so. This is because a wider theoretical context might invalidate the cognitive value of these propositions, as stated by James Anderson
(2005, pp. 64–65
), explaining Cornelius Van Til’s argument: “if no one has comprehensive
knowledge of the universe, then no one can have any
knowledge of the universe’ because ‘unless one knows everything
about universe, the interrelatedness of the university means that whatever reasons or grounds one has for one’s beliefs the possibility remains of some fact coming to light that radically undermines those reasons or grounds.”
As you can see, by constantly revising their findings, scholars gain a continually improved level of knowledge in the course of the development of science (although not necessarily in a linear manner). The ideal boundary (or apogee or horizon) to this development is ultimate or perfect knowledge as the only true, complete, and relevant par excellence one. This is the knowledge we would ascribe to the omniscient being, i.e., God or someone who cannot be distinguished from God from the epistemological point of view.
Do scholars, in their pursuit of real (revisable and limited) knowledge, refer to perfect knowledge in any way? Yes, at least unconsciously. Firstly, because they seek this knowledge by constantly revising, complementing, and deepening the currently available or grasped knowledge. Secondly, because they, nolens volens
, compare the actual state of their real knowledge with the presumed ideal knowledge. Thirdly, because in the aforementioned processes of seeking and comparing, they objectify their knowledge—without these processes, it would appear as unjustified, i.e., as merely a set of opinions about the world, formulated by certain people at a given time. In other words, only in the context of the ultimate or perfect knowledge horizon can current knowledge be assessed as an approximation of what the world is. Our actual knowledge holds value only in relation to perfect knowledge, or (to use words from another context) in relation to “a God’s eye point of view” (Putnam 1990, p. 23
I.3. Idealization?—Premises (3a) and (4)
Does, or can, this “God’s eye point of view” knowledge actually exist? Proponents of the negative answer to this question argue that perfect knowledge
is just an idealization of all the actual knowledge we can possess. This idealization is formulated to understand the development of science better. Similarly, as it would be put by Immanuel Kant
(1998, A 692, B 720, p. 617
), it can be argued that perfect knowledge
is only a regulative idea of our cognition—one that inspires us to continuously improve our search for the unity or the completeness of knowledge.
Opponents of this approach, with whom we identify, will note three facts.
Firstly, the concept of perfect knowledge is not only an idealization in the sense of a simplified representation of an object. It is not a kind of average human knowledge but something much more. Indeed, this concept not only gives an account of what our scientific knowledge is but also signifies the presumed (ultimate) goal of our scientific endeavour. After all, we have established that scholars, whether consciously or unconsciously, seek perfect or ultimate knowledge.
Secondly, note that typical idealization, as used in science—let us call it factual idealization—is something of a springboard to a more complete knowledge of the subject through concretization. This is not the case, however, with perfect knowledge. Such knowledge, if someone possesses or will possess it, does not require, contrary to factual idealization, any concretization. It is, rather, as you will see later, a hypothetical ending point and, at the same time, a normative reference point for any scientific endeavour. In addition, it is only in the context of the ideal standard of perfect knowledge that we can understand the significance of the factual idealization better. For, when we use concretization to get closer to the truth, we acknowledge that our goal is to achieve perfect or ultimate knowledge.
Thirdly, the aforesaid fact of seeking perfect knowledge is connected with the existential involvement of scholars. One cannot seek something without any form while believing that it really exists. If so, perfect knowledge
is not a regulative idea but a prerequisite for the validity of scientific endeavour, one that is socially significant and, at the same time, engages its participants. (See the biographies of eminent scholars, driven by scientific passion, devoting their whole life to science.) To paraphrase Kant
(1998, A 811, B 839—A 813, B 841, pp. 680–81
), just as God is not a regulative idea of morality but a prerequisite for (or a necessary condition of) its fulfilment, God (as the subject of perfect knowledge) is not so much the regulative idea behind cognition but the precondition for its realization.
Indeed, there is an analogy between practicing morality and practicing science. It can be shown by paraphrasing Linda Zagzebski’s conclusion concerning the relationship between God and morality. As put by Zagzebski
(1987, p. 294
), “it is not rational to try to be moral unless it is rational to believe that the attempt has a reasonable chance for success. But it is not rational to believe success unless one believes there is a factor which explains how. A providential God is such a factor”.
We would paraphrase Zagzebski saying that it is not rational to try to do science unless it is rational to believe that the attempt has a reasonable chance for success. However, doing science cannot be successful unless there is a factor that makes success possible. Perfect knowledge (possessed by God) is such a factor.
Fourthly, accepting the reality of perfect knowledge allows us to understand certain important phenomena associated with scientific activity. Jan Łukasiewicz, a famous Polish logician and philosopher of science, the practical dimension of which he knows from his own experience with his work in logic, includes in this group the sense of pressure or obligation imposed by the ideal of truth, the admiration for its grandeur, and the faith in its binding power. He wrote (Łukasiewicz 1970, p. 249
“In concluding these remarks, I should like to outline an image which is connected with the most profound intuitions which I always experience in the face of logistic. That image will perhaps shed more light on the true background of that discipline, at least in my case, than all discursive description could. Now, whenever I work even on the least significant logistic problem, for instance, when I search for the shortest axiom of the implicational propositional calculus, I always have the impression that I am facing a powerful, most coherent and most resistant structure. I sense that structure as if it were a concrete, tangible object, made of the hardest metal, a hundred times stronger than steel and concrete. I cannot change anything in it; I do not create anything of my own will, but by strenuous work I discover in it ever new details and arrive at unshakable and eternal truths. Where is and what is that ideal structure? A believer would say that it is in God and is His thought.”
According to this testimony, scientific work is sometimes realized in a vague, yet distinctly sensed, prospect of ideal knowledge or structure, which is normative in nature in relation to our current cognitive actions.
I.4. A Humanist Utopia?—Premises (3b) and (5)
One might say that opponents of the conception of perfect knowledge, as an idealization, do not have to take a theist approach. After all, they can subscribe to the solution providing for a humanist or transhumanist quasi-deification. According to this approach, perfect knowledge is neither a fictional equivalent of idealization nor the knowledge currently possessed by God, but potential knowledge that can (or really will) be acquired by mankind in the future.3
In a sense, this is more or less what Steven Hawking
(1988, p. 155
) meant when he wrote, “Ultimately, however, one would hope to find a complete, consistent, unified theory that would include all these partial theories as approximations.”
This brilliant physicist recognised the number of problems the idea of the “ultimate theory of the universe” presents. He did not, however, rule out its (human!) realization, which would be crowned by “the ultimate triumph of human reason—for then we would know the mind of God” (Hawking 1988, p. 185
Regardless, if the sentence cited above, the last sentence of the famous Brief History of Time, is construed literally, or as a rhetorical device, it reveals the alternative option of humanist, scientific eschatology. In this conception, mankind, or a similar community, is expected to achieve (probably via the appropriate computational devices) a state which nearly achieves perfect knowledge. We do not intend to explore the technical aspects of the possibility of achieving such knowledge. What we will do here is give two fundamental reasons which, in our opinion, eliminate the possibility of autonomous acquisition of such knowledge by finite beings.
Firstly, human minds (even supplemented with electronic computational devices) are finite. However, finite minds cannot believe all—infinite in number!—true propositions. Thus, human minds (or human extended minds) cannot have perfect knowledge. As Roman Ingarden
(2016, p. 221
) put it, each object of cognition is distinguishable by the “transcendence of the plenitude of being” with respect to the cognitive subject of consciousness.4
In other words, due to the infinite number of its properties and relations it enters or can enter, no element of the world can be exhausted in cognitive terms through a finite number of cognitive acts.
Secondly, let us suppose that mankind has a scientific theory that is assumed to cover all the primary or ultimate frameworks governing the world. Let us also assume that this theory allows us to generate a huge number of true propositions about the world, which produce a relatively exhaustive picture of the world. Even if this situation was the case, it can hardly be concluded that scholars could have any proof of the completeness and incontrovertibility of their ultimate theory.5
Various, somewhat manageable but irrefutable, random circumstances of our cognition always allow either the possibility of a mistake or the appearance of new data which might prompt us to change our theory (by facilitating even more primary frameworks to be discovered or postulated). Such knowledge—one we need to label as “we do not know (for sure) that we know”—is not, in fact, perfect knowledge.
I.5. God as the Subject of Perfect Knowledge—Premises (3c) and (6)
Please note that, in order to satisfy the two above-mentioned reservations, knowledge would have to be comprehensive, i.e., exhaustive. In other words, perfect knowledge should encompass all its objects in their entirety, including all their relations, stages, and complete history.
Comprehensive knowledge is not possible for individual human beings, since they are limited in terms of both time and space. It also does not seem possible for larger scientific communities that embrace greater—yet still limited in terms of their existence and scope of investigation—areas of space and time. Let us remember that comprehensive knowledge requires that an object not only be captured in all its possible aspects but also throughout its whole existence and across all its temporal and spatial relations. Therefore, it needs to be noted that comprehensive knowledge is indeed different from even the most adequate knowledge achieved in the heretofore available scientific processes.
Given the above, it needs to be further noted that the subject of perfect knowledge must have certain qualities. First of all, it must transcend the world in terms of time and space, instead of being part of it or being limited by any of its areas. It also needs to have an absolute, unbiased cognitive access to the world and each of its elements. Such a subject can hardly be associated with any other being than God, since, as argued by Dummett
(2006, p. 96
), only God “has no particular point of view, no location in
the world, no perspective contrasted with other perspectives. He knows, not by the effect of objects or events upon His perceptual equipment, but by His comprehension of all truth.”
In other words, the subject of perfect knowledge needs to be identified with an epistemically perfect being. The necessary ontic conditions for this epistemic perfection seem to be transcendence, infinity, and primacy over the world (also in the sense of ruling it). To call such an epistemically and ontically perfect being God would certainly be consistent with our philosophical and cultural tradition. Any detailed description of this idea of God and His relation to the world is, however, outside the scope of the proposed argument.
The above-mentioned argument, as developed here, meets with a number of objections which allowed us to recognise its potential flaws. Four of them will be addressed below.6
II.1. The First Objection: Platonism
Premise (3) enumerates three possibilities, including idealization, (trans)humanism, and theism. If it is complete, then the major veins of the argument are premises (4) and (5), which reject the first two of the above-mentioned possibilities or options. However, is it indeed complete? The objection can be made that in our deliberations we have overlooked one important, and well-known throughout the history of philosophy, option—namely, Platonism. According to it, ultimate or perfect knowledge is a collection of all ideally existing (and at least potentially true) propositions (in the logical sense). They exist in themselves outside space-time and minds as supra-subjective or autonomous entities and need not be recognized in order to exist.
However, there are at least two reasons against this conception.
Firstly, whoever believes in the existence of any knowledge without minds falls into a trap of abstraction separating true justified propositions from their bearers. It is so because we are not acquainted with any knowledge without its conscious carrier and we have no basis to state that such knowledge is possible. Maybe we can use the term knowledge
in a different way, but it contradicts our linguistic intuitions.7
Secondly, knowledge is not only a collection of several propositions but also a set of (also successive) interrelations between them and operations performed on them. It is difficult to comprehend these relations and operations without a unifying factor, i.e., the mind.
As you can see, even taking Platonism, or related approaches, into account will not change the outcome of our deliberations. Just as (trans)humanism does not give a sufficient ground for perfect knowledge, Platonism does not give any ground for it.
II.2. The Second Objection: The Parody of the Perfect Golf Player
Advocates of the conception of perfect knowledge as an idealization could formulate a parody of our argument (after the fashion of the famous Gaunilo’s parody of Anselm’s ontological argument). It goes as follows: if the goal of practicing science presupposes the existence of the subject of perfect knowledge, then—to use the example of a sports game—the goal of the game of golf presupposes the existence of the perfect golf player, i.e., the golf player who makes a hole-in-one in every actual or every possible game. In other words, to assume our conception of perfect knowledge is to open the door for various fictitious beings, which can be considered as the bearers or carriers of ideal goals of different human activities.
There are two approaches to addressing this parody. The first initially accepts its conclusion. This acceptance recognizes our argument and the golf argument
as (not equivalent) members of a vast family of arguments “from ideals” or “from different degrees of perfection.” The purpose of these arguments is to prove the existence of a perfect being from the point of view of different kinds of perfection, i.e., the perfect knower, the perfect golf player, etc. The economy of thinking requires, however, that concepts of these entities must be unified into one, to postulate a single referent. In addition, in line with this economy, the features ascribed to this referent need to be reduced to positive essential features or, in scholastic terms, to pure perfections, i.e., those whose exemplification is definitely better than the lack thereof (or than the exemplification of their opposites). Under this approach, God appears as the perfect subject of pure perfections, including superior cognition and action. This clearly demonstrates that God is the subject of perfect knowledge (as a pure perfection) but not the subject of perfect golf playing (which is not any pure perfection).8
There is no doubt that knowledge (and perfect knowledge) is absolutely better than ignorance or the lack of knowledge, but golf playing is not better than its lack.9
The second strategy is to directly note the differences between (perfect) knowledge and (perfect) golf playing. Let us mention three such differences.
Firstly, both perfections differ in terms of their existential significance. The game of golf, contrary to knowledge, has no considerable importance for the lives of persons in general. As a human being, you can live without playing golf, but without knowledge, you cannot.
Secondly, the concepts of both perfections are formed differently. The concept of a perfect golfer is formulated on the basis of a single case of a player hitting the target. Via the multiple mental iterations of this empirical and single occurrence, we arrive at the concept of an individual who holes out in every possible game. This is not the case with perfect knowledge. It seems that this concept is provided to us through the simple intuition of such knowledge, which serves to recognize, define, and assess imperfect knowledge against this initial intuition. In other words, the first concept, as a summative one, is a derivative term, while the latter is a primitive term.
Thirdly, both perfections differ in terms of their nature. Perfect or unlimited knowledge, as opposed to unlimited golf playing, is metaphysically possible. It is possible to think of an entity that knows everything without any limits. On the other hand, it is not possible to play golf (a physical game in nature!) in an unlimited manner. We are always confined to the potential of our bodies and the laws of physics.
The above-mentioned differences or a reference to the economy of thinking mentioned in the first strategy lead to the conclusion that the parody of the perfect golfer is not a serious threat to the soundness of our argument.
II.3. The Third Objection: Ordinary Scientists
It could be said that our argument mistakenly identifies the goal of science with perfect knowledge. After all, we know that scientists set themselves different goals, and ordinary scientists, working in the phase of so-called normal science, perform ordinary activities such as collecting empirical data and integrating them into scientific schemes (in accordance with the current paradigm). Thus, there is no place for perfect knowledge in the description of real scientific practice.
Responding to this objection, it is necessary to distinguish the personal goals of scientists and research communities from the intrinsic (or immanent) and objective goal of the very science they practice. Every cognitive undertaking, regardless of the incidental goals of its implementers, aims at truth. This is even more so with science as a particularly sophisticated cognitive endeavour that seeks complete and essential truth, i.e., perfect knowledge. Although scientists are imperfect in various respects, the inner content of their endeavours by its very nature tends towards this knowledge.
If someone, for philosophical reasons, does not agree to operate with the notion of immanent goal, we can formulate our argument in terms of a deep human aspiration to perfection. This aspiration is realised in various fields (see the answer to the previous objection), and especially in the field of science. If this aspiration is real, then (acknowledging the other premises of our argument) it cannot be explained without reference to a perfect (and in particular, cognitively perfect) God. One might even say that the aspiration to perfection hidden in us reveals to us the hidden God.
II.4. The Fourth Objection: The Gap between Divine and Human Knowledge
One can say that our argument is both theoretically and practically inconclusive. Theoretically inconclusive because it is impossible to proceed from premises concerning human knowledge to a conclusion concerning divine knowledge; the latter is, after all, incomparably superior to the former. Practically inconclusive because the human aspiration to divine knowledge can never be realised. In that case, even if divine knowledge exists, since it can never be attained by us, it has no relevance to us.
To answer the first part of the objection, we propose to distinguish between the extent of knowledge and the mode of access to knowledge. In terms of access to knowledge, divine access to knowledge is incomparably better than human access. As we have already written, God grasps all possible truths in one intuitive comprehension, whereas man, in laborious work, arrives at discursive knowledge of only some of them. On the other hand, the set of all truths known by God, though vast, is definite. Thus, we can say that the set of truths known by humans is a subset of it.
As for the second part of the objection, it should be noted that our argument does not assume that humans will ever (in a natural way) attain perfect knowledge. We ruled out this possibility in premise (5). Our argument only assumes that our knowledge remains in relation to perfect knowledge. This relation, on the other hand, does not exclude that in the eschatological perspective humans will receive the grace of a special insight into divine knowledge—an insight that exceeds natural human faculties but does not exhaust divine knowledge. As Plantinga would say, “then in our thinking we would literally be thinking God’s thoughts after him” (Plantinga 2007, p. 211
III. General Conclusion: The Indispensability of the Concept of God
Undoubtedly, not everyone will be convinced by our responses to the aforementioned objections. Indeed, not everyone will agree to dismiss conceptions alternative to our conception of perfect knowledge. Please note, however, that even if you accept the idealization (3a) or humanist (3b) conception of perfect knowledge, you must, even though you do not acknowledge the existence of God, use the concept of a being similar to Him. In the first case, it is the irreversibly empty (yet cognitively useful) concept of the subject of ultimate knowledge, while, in the second, the concept of such a subject (or a similar one), which will become non-empty in the future or now, is potentially non-empty. This demonstrates that, regardless of how the issue of the existence of its referent is addressed, the concept of God (or a similar one) is needed in any wider (metaphysical) contemplation of scientific practice.
To conclude, let us quote a fragment from a famous Polish handbook of the methodology of science, wherein Kazimierz Ajdukiewicz
(1974, p. 187
) argues that:
“It has been said that the procedures used by scientists in a given discipline look so, in a synthesizing approach, as if those scientists were striving towards a goal. This is not to say that those scientists always realize that. Yet they act in the way they would act if they realized what their goal is. If they act so without realizing clearly what their goal is, then it may be said that they are striving towards that goal unconsciously. One of the tasks of the methodologists is to identify those goals towards which scientists working in a given field strive, whether consciously or unconsciously”.
In our opinion, this, often unrealized, objective or goal is (in a wider or deeper perspective of metaphysics of science) to pursue ultimate or perfect knowledge. The understanding of this concept, requires, however, a reference to the concept of God or a similar entity. Given the above, it seems reasonable to venture a hypothesis that scholars, even those who are declared atheists, behave in their work as if they believed in God or in some sort of His substitute.
In a sense, this hypothesis is confirmed by Richard Rorty
(1999, p. 38
), a thinker who is far from theism. In his opinion, “there would only be a ‘higher’ aim of inquiry called ‘truth’ if there were such a thing as ultimate justification—justification before God, or before the tribunal of reason, as opposed to any merely finite human audience.”
As you can see, Rorty leaves us with a dilemma. We can either accept the “higher aim of inquiry”, thus acknowledging some form of theism or quasi-theism, or renounce it for the benefit of epistemological relativism, which de facto undermines the very foundations of science itself. Rorty opted for the latter alternative. We, in turn, due to the respect we have for the authority of scholars and the admiration of scientific achievements, have chosen the former.